Animation UsingMotion Capture

Edmond C. PrakashEdmond.Prakash@beds.ac.uk

Overview

• What is MoCap?

• Why do we need Motion Capture?

• Why is MoCap difficult?

• If you are given RM250,000 to setup a MoCap facility what will you do?

MoCap Outline

• Introduction - Capture of MoCap Data• Computational Issues

• Data Format• Data Base • MoCap Editors• Motion Graphs• Retargetting

• Real-time MoCap for Games• Summary

Motion Capture

Motion Capture

What type of animation can be synthesized from motion capture?Can capture behavior far too complicated to model by hand.– Dance– Martial Arts– Sports– Acrobatics, other stylized motion

Motion Capture Applications. Who needs it?

• Movie Industry and Game Industry

• Animated Movies/Game development often calls for enormous quantities of motion capture.

• Goal: – Increase animation production capabilities. – To achieve higher quality.

MoCap Devices• Whole Body (Light/Sensors/Emitters/LEDs)

– Ascension Technology Corp www.ascension-tech.com– Meta Motion www.metamotion.com – Motion Analysis Corp www.motionanalysis.com – Mova Inc www.mova.com– PhaseSpace Inc www.phasespace.com – PTI Phoenix Technologies www.ptiphoenix.com – Polhemus ww.polhemus (magnetic) www.polhemus.com – Vicon www.vicon.com

• Hand and Fingers– Immersion Corp (wireless glove for hand and fingers)

www.immersion.com/3d – Measurand Inc (wireless data glove for handsand fingers)

www.measurand.com

• Face– Image Metrics www.image-metrics.com MOCAP Data business

MoCap Outline

• Introduction - Capture of MoCap Data

• Computational Issues • Data Format• Data Base • MoCap Editors• Motion Graphs• Retargetting

• Real-time MoCap for Games• Summary

Motion Capture – Computational Issues

• The quick and dirty answer to animation?– Well, it may be “dirty”, but it’s not quick!

• It is hard to get:

– Marker position problem

– Joint center estimation

– Body suits, clothing, slippage

• It’s Hard To Use – The pipeline can be complex– Who did we capture?

– To whom (or what!) can it be applied?

• Motion capture– Filtering, editing, retargeting, stitching,…

MoCap Outline

• Introduction - Capture of MoCap Data

• Computational Issues • Data Format• Data Base • MoCap Editors• Motion Graphs• Retargetting

• Real-time MoCap for Games• Summary

MoCap Data Formats

• Biovision and Acclaim are two of the most popular and standard motion capture file formats.

• Biovision BVH: Biovision Hierarchical motion capture data (.bvh). A .bvh file is a text file. The BVH file format is a joint based format and it is the most popular format.

• Acclaim ASF and AMC: The Acclaim file format stores the top-level skeletal information in a .asf file and multiple motion capture datasets in .amc files. Acclaim is a bones based file format.

• Most studio houses that do their own commercial animation have their own type of motion capture technique.

BVH Skeleton• Hips

– LeftHip• LeftKnee

– LeftAnkle

– RightHip• RightKnee

– RightAnkle

– Chest• LeftCollar

– LeftShoulder» LeftElbow LeftWrist

• RightCollar– RightShoulder

» RightElbow RightWrist

• Neck– Head

BVH File formathttp://www.cs.wisc.edu/graphics/Courses/cs-838-1999/Jeff/Example1.bvh

HIERARCHYROOT Hips{

OFFSET0.000.000.00CHANNELS 6 Xposition Yposition Zposition Zrotation Xrotation YrotationJOINT Chest{

OFFSET0.004.210.00CHANNELS 3 Zrotation Xrotation YrotationJOINT Neck{

OFFSET0.0017.080.00CHANNELS 3 Zrotation Xrotation YrotationJOINT Head{

OFFSET0.003.970.00CHANNELS 3 Zrotation Xrotation YrotationEnd Site{

OFFSET0.003.230.00}

}}

…

BVH File format …2JOINT LeftCollar{

OFFSET 1.02 14.86 1.71CHANNELS 3 Zrotation Xrotation YrotationJOINT LeftUpArm{

OFFSET 5.45 0.00 0.00CHANNELS 3 Zrotation Xrotation YrotationJOINT LeftLowArm{

OFFSET 0.00 -11.29 0.00CHANNELS 3 Zrotation Xrotation YrotationJOINT LeftHand{

OFFSET 0.00 -8.34 0.00CHANNELS 3 Zrotation Xrotation YrotationEnd Site{

OFFSET 0.00 -6.13 0.00}

}}

}}

Second part of the .bvh file:

MOTIONFrames: 2Frame Time: 0.033333 ! 30 frames a second

8.03 35.01 88.36 -3.41 14.78 -164.35 13.09 40.30 -24.60 7.88 43.80 0.00 -3.61 -41.45 5.82 10.08 0.00 10.21 97.95 -23.53 -2.14 -101.86 -80.77 -98.91 0.69 0.03 0.00 -14.04 0.00 -10.50 -85.52 -13.72 -102.93 61.91 -61.18 65.18 -1.57 0.69 0.02 15.00 22.78 -5.92 14.93 49.99 6.60 0.00 -1.14 0.00 -16.58 -10.51 -3.11 15.38 52.66 -21.80 0.00 -23.95 0.00

7.81 35.10 86.47 -3.78 12.94 -166.97 12.64 42.57 -22.34 7.67 43.61 0.00 -4.23 -41.41 4.89 19.10 0.00 4.16 93.12 -9.69 -9.43 132.67 -81.86 136.80 0.70 0.37 0.00 -8.62 0.00 -21.82 -87.31 -27.57 -100.09 56.17 -61.56 58.72 -1.63 0.95 0.03 13.16 15.44 -3.56 7.97 59.29 4.97 0.00 1.64 0.00 -17.18 -10.02 -3.08 13.56 53.38 -18.07 0.00 -25.93 0.00

Other MoCap Data Formats

• Biovision and Acclaim are two of the most popular and standard motion capture file formats.

• Biovision BVH: Biovision Hierarchical motion capture data (.bvh). A .bvh file is a text file. The BVH file format is a joint based format and it is the most popular format.

• Acclaim ASF and AMC: The Acclaim file format stores the top-level skeletal information in a .asf file and multiple motion capture datasets in .amc files. Acclaim is a bones based file format.

• Most studio houses that do their own commercial animation have their own type of motion capture technique.

• Formats, Clips, Motion Library

MoCap Outline

• Introduction - Capture of MoCap Data• Computational Issues

• Data Format• Data Base • Mocap Editors

• Motion Editing• Motion Graphs• Retargetting

• Real-time MoCap for Games• Summary

A simple motion graph

Walk2

Walk1

Run2/Jog

Jog/Walk1

Run1

Run2

Walk1/Walk1

Run2/Run2

Jog

Walk1 Walk2 Jog Run1 Run2

Original motion data

Walk2/Walk1

MoCap Outline

• Introduction - Capture of MoCap Data• Computational Issues

• Data Format• Data Base • MoCap Editors

• Motion Graphs

• Retargetting• Multiple Character Synchronisation

• Real-time MoCap for Games• Summary

Editing of MoCAP Data - Retargeting Motion to New Characters

Simple Try 1 (Space Problem)• Just reuse parameters of original motion in

new motion with different character

• hmm … something’s not quite right …

Simple Try 2 (Time Problem)

• Need to meet constraints, for instance foot touches floor

• No problem, just use inverse kinematics to establish constraints each frame

… better, but not very smooth …

Multiple Character Synchronisation

• Collaborative and adversarial interaction among multiple characters usually requires synchronization in both space and time.

• Transformation handles are required to edit the animation specified using a motion graph are:-to specify absolute/relative spatial location, direction, time, duration, and synchronization of multiple characters.

3 Character Synchronisation

Two synchronized tracks: Time warping

The original uniform time sequences at the top are interactively manipulated to generate smooth, synchronized time-warps. The user transforms only one track and the other track is warped accordingly.

Discrete motion editing

(Top) input walking motion.

(Middle) Path editing necessarily entails longer strides.

(Bottom) Discrete motion editing introduces extra steps to cope with excessive stretching.

MoCap Outline• Introduction - Capture of MoCap Data• Computational Issues

• Data Format• Data Base • MoCap Editors• Motion Graphs• Retargetting• Multiple Character Synchronisation

• Real-time MoCap for Games• Summary

Motion-Captured

Gameplay

• Scenario 1: Real-time allows the game player to choose from pre-created moves (eg. raise an arm), to control the character's moves in real-time.

• Scenario 2: Players walk about on the floor, thus controlling their characters' actions.. If players want their characters to raise an arm, they do so.

http://www.torquepowered.com/community/blogs/view/12041/1

References• Barbara Robertson, Big Moves, Computer

Graphics World, November 2006• Jeff Lander, Working with Motion Capture

Formats, Game Developer, January 1998• M. Gleicher, Retargetting motion to new

characters. Computer Graphics (Proceedings of SIGGRAPH 98), 32:33–42, July 1998.

• Kim, Manmyung and Hyun, Kyunglyul and Kim, Jongmin and Lee, Jehee, Synchronized multi-character motion editing, ACM Trans. Graph., vol. 28, no. 3, 2009, pages 1—9.